BACKGROUND: Reduced heart rate variability (HRV) is associated with a poor outcome in patients with sinus rhythm (SR) or atrial fibrillation (AF). However, cut-off points for HRV measures differ between SR and AF. We hypothesized that a global index of 24-hour HRV based on evaluation of scatterplot would describe HRV irrespective of cardiac rhythm. METHODS: 407 patients with ischemic heart disease (317 male, 90 female, mean age 57 ± 9 years) were studied. 331 patients had SR and 76 patients had AF. 24-hour ECGs were recorded, and standard HRV indices were calculated. Scatterplots was used to determine the HRV fraction (HRVF, %). HRV measures were compared in respect to left ventricular ejection fraction (LVEF £ 35% or 〉 35%). RESULTS: Standard HRV measures were higher in AF-patients despite the mean RR interval was lower. In patients with LVEF £ 35%, standard HRV indices were lower in SR group, in AF group only SDNN and RMSSD were reduced. The HRVF was comparably reduced (SR 39.3 ± 15.3%, AF 37.3 ± 17.9%). In patients with LVEF 〉 35%, HRVF did not differ between SR (47.2 ± ± 10.5%) and AF (46.1 ± 12.1%). The HRVF correlated with SDNN and SDANN (~0.85) in SR. Correlations were weaker in AF (~0.6). Standard HRV indices and HRVF showed similar relations with LVEF, but only in AF at the same range. CONCLUSIONS: The HRV fraction allows for HRV evaluation irrespective of cardiac rhythm. The index elicited a similar dependence of HRV on left ventricular function in SR and AF.
BACKGROUND: Reduced heart rate variability (HRV) is associated with a poor outcome in patients with sinus rhythm (SR) or atrial fibrillation (AF). However, cut-off points for HRV measures differ between SR and AF. We hypothesized that a global index of 24-hour HRV based on evaluation of scatterplot would describe HRV irrespective of cardiac rhythm. METHODS: 407 patients with ischemic heart disease (317 male, 90 female, mean age 57 ± 9 years) were studied. 331 patients had SR and 76 patients had AF. 24-hour ECGs were recorded, and standard HRV indices were calculated. Scatterplots was used to determine the HRV fraction (HRVF, %). HRV measures were compared in respect to left ventricular ejection fraction (LVEF £ 35% or 〉 35%). RESULTS: Standard HRV measures were higher in AF-patients despite the mean RR interval was lower. In patients with LVEF £ 35%, standard HRV indices were lower in SR group, in AF group only SDNN and RMSSD were reduced. The HRVF was comparably reduced (SR 39.3 ± 15.3%, AF 37.3 ± 17.9%). In patients with LVEF 〉 35%, HRVF did not differ between SR (47.2 ± ± 10.5%) and AF (46.1 ± 12.1%). The HRVF correlated with SDNN and SDANN (~0.85) in SR. Correlations were weaker in AF (~0.6). Standard HRV indices and HRVF showed similar relations with LVEF, but only in AF at the same range. CONCLUSIONS: The HRV fraction allows for HRV evaluation irrespective of cardiac rhythm. The index elicited a similar dependence of HRV on left ventricular function in SR and AF.